Labour market institutions and wage setting: evidence for OECD countries

Abstract

The main channel through which labour market institutions are supposed to work in affecting unemployment is through their effects on the key parameters of the wage curve. In particular, labour market institutions may have both a direct wage push (or level) effect, i.e. change the level of the real wage for any given level of the unemployment rate and productivity, and an indirect slope effect, i.e. change the responsiveness of the real wage to the unemployment rate. The question this article addresses is whether there is any evidence that these transmission mechanisms were at work in a group of 20 Organization for Economic Co-operation and Development (OECD) countries over the period 1960 to 1999. The analysis is accomplished in two steps. Pooled Mean Group (PMG) estimates of a wage equation including unemployment, productivity and a set of wage push institutions are first obtained, allowing only a subset of institutional coefficient to be homogeneous, while leaving the unemployment and other coefficients free to differ across countries. The country specific estimates of the unemployment coefficients are then used to investigate whether and to what extent cross-country heterogeneity in the estimated wage response to unemployment is related to institutional differences. The results support the existence of significant wage push effects of union density and benefit replacement rates, and of significant slope effects of benefit replacement rates, benefit duration and employment protection. A more generous unemployment benefit structure is found to lower the wage responsiveness to unemployment, while higher employment protection, contrary to what one expects, is found to enhance it. No significant level and slope effects are found for the tax wedge and bargaining coordination.